- Email: [email protected]
Nasal Carriage of Methicillin-Resistant Staphylococcus aureus in Vascular Surgery
Papers Presented to the French Vascular Surgery Society Nasal Carriage of Methicillin-Resistant Staphylococcus aureus in Vascular Surgery Virginie Morange-Saussier, MD,1 Bruno Giraudeau, PhD,2 Nathalie van der Mee, PhD,1 Patrick Lermusiaux, MD,3 and Roland Quentin, PhD,1,4 Tours, France
The purpose of this study was to determine the prevalence of nasal carriage of methicillinresistant Staphylococcus aureus (MRSA) and to define risk factors allowing identification of high-risk patients for MRSA nasal carriage at admission to the vascular surgery unit. From March 23, 2004 to July 13, 2004, screening for nasal carriage of MRSA was conducted at admission to the vascular surgery unit and 1 week thereafter. To analyze risk factors for MRSA nasal carriage at admission to the vascular surgery unit, a case-control study was carried out in patients presenting colonization at the time of admission. A total of 308 patients underwent nasal screening for MRSA. Thirteen were colonized with MRSA (nine at admission and four acquired), i.e., 4.2% of patients. Methicillin-susceptible Staphylococcus aureus (MSSA) was found in 11.4% of patients who underwent screening. Six patients with MRSA infection were identified during the study period. The two patients who acquired infection were colonized at the time of admission. Arrival from another health-care facility and from another department was a significant risk factor for carriage of MRSA. The prevalence of nasal carriage in vascular surgery was 4.2%. Nasal screening is highly cost-effective since 60% of MRSA carriers were undetected using diagnostic specimens alone. French recommendations issued for cardiac and orthopedic surgery by the consensus conference on preoperative management of infectious risk on March 5, 2004, should be extended to vascular surgery.
INTRODUCTION Antibiotic multiresistant bacteria (AMB) are a major cause of nosocomial infection (NI) in
1 Department of Bacteriology and Hospital Hygiene, Trousseau University Hospital, Tours, France. 2 INSERM CIC202, Franc¸ois Rabelais University and Regional University Hospital, Tours, France. 3 Vascular Surgery Department, Trousseau University Hospital, Tours, France. 4 EA 3854 ‘‘Bacteria and Materno-Fetal Risk,’’ Microbiology Department, University of Tours School of Medicine, Tours, France. Presented at the Twentieth Annual Meeting of the Socie´te´ de Chirurgie Vasculaire de Langue Franc¸aise, May 30-31, 2005, Lyon, France. Correspondence to: Virginie Morange-Saussier, MD, Bacteriology and Virology Department, Bretonneau Regional University Hospital Center, 37044, Tours, Cedex 1, France, E-mail: [email protected]
Ann Vasc Surg 2006; 20: 767–772 DOI: 10.1007/s10016-006-9088-x Ó Annals of Vascular Surgery Inc. Published online: November 4, 2006
France. The usual definition of NI is a hospitalacquired infection that was absent at the time of admission or developed within 48 hr after admission. Surgical site infections (SSIs) are considered nosocomial if they occur within 30 days after the intervention or, in cases involving implantation of a prosthesis or other device, within 1 year. Staphylococcus aureus is one of the two main bacterial species causing NI. Methicillin-resistant S. aureus (MRSA) is an AMB accounting for 5-10% of bacteria isolated during NI and 50% of S. aureus-related SSIs overall for all types of surgery.1 The multiresistance of these strains is defined as resistance to three antibiotics other than ß-lactamines, which limits the choice of antimicrobial therapy. Resistance of S. aureus to methicillin remains widespread in France and is involved in over 30% of all infections and 50% of NIs alone. High prevalence rates have also been documented in the United States (50%). A 767
768 Morange-Saussier et al.
previous study in 1990 showed that European countries could be divided along south and north lines with regard to the proportion of MRSA present in S. aureus strains, with Italy, France, and Spain showing around 30% and Denmark, Sweden, and the Netherlands showing around 2%.1 The lowest levels were observed in the Netherlands, with <1% of S. aureus found in clinical specimens being identified as MRSA.2 The 2001 national study of the prevalence of NI in France confirmed the high prevalence of AMB, i.e., 6.9% nationwide and 6.4% at the Tours University Hospital Center. MRSA accounted for 64% of S. aureus strains isolated from NI nationwide and 38% isolated at the Tours University Hospital. Whether symptomatic or not, infected or colonized carrier patients represent reservoirs for MRSA. They are a potential source of spreading and a major cause of epidemic MRSA outbreaks. Wound and urine infected with MRSA are important reservoirs. The nares, a preferential site of S. aureus colonization, are also an important reservoir. S. aureus colonizes the nares in 10-40% of individuals with no hospital contact. In the hospital setting and during epidemic outbreaks, up to 7% of the medical staff may be MRSA carriers.3 Transmission from carrier patients occurs primarily through the hands of health-care workers. However, transmission can also occur by contamination of the surface of inanimate objects such as stethoscopes and blood pressure cuffs. Screening and application of preventive measures for MRSA have been shown to be effective, especially in intensive care units. However, there are few data on other health-care settings, in particular for vascular surgery units.3 In this regard, it should be pointed out the risks associated with nasal carriage of MRSA have been confirmed for cardiac surgery, which presents conditions much like those observed in vascular surgery. A special feature of vascular surgery is to treat both septic and aseptic patients. The proximity between clean and contaminated patients raises a high risk of cross-transmission and therefore of catastrophic life-threatening prosthetic infection. By detecting MRSA carriers at admission, nasal screening may allow immediate implementation of control measures to prevent spreading. The purpose of this study was to establish the prevalence of MRSA carriage at admission to the vascular surgery unit and to evaluate the impact of eight risk factors for nasal carriage of MRSA in vascular surgery.
Annals of Vascular Surgery
MATERIALS AND METHODS Population This study included all patients hospitalized longer than 24 hr in a 19-bed vascular surgery unit over the 4-month period between March 23 and July 13, 2004. Microbiology Nurses took swab specimens from both of the nares of each patient at admission and 1 week thereafter. Patients who were readmitted were included. On the first day, nare specimens were seeded according to conventional culture technique on a selective Mueller-Hinton medium supplemented with tobramycin at 10 mg/L and on 5% nonselective blood agar gelose (bioMe´rieux, Marcy lÕEtoile, France). Gelose was then incubated for 24 hr at 37°C. On the second day, genus was diagnosed based on colony appearance, presence of a catalase, and characteristic microscopic morphology after gram staining (clustering of gram-positive cocci). S. aureus species identification was based on detection of bound coagulase by latex agglutination testing (PastorexÒ Staph-Plus; Bio-Rad, Ivry-sur-Seine, France). Speciation was confirmed biochemically using an automated Rapid ID32 STAPH system (bioMe´rieux). On the third day, susceptibility to methicillin was measured for each S. aureus strain (manual diffusion technique using a cefoxitin disk on gelose according to the recommendations of the Antibiogram Committee of the French Society of Microbiology in 2004).4 If necessary, methicillin resistance could be confirmed using either an oxacillin 5 mg disk incubated at 30°C for 24 hr, a rapid latex agglutination test (SlidexÒ MRSA detection, bioMe´rieux) to detect the PLP2A protein causing methicillin resistance, or molecular biology techniques to detect MecA coding for the PLP2A protein. MRSA from cultures ordered for diagnostic purposes by the clinicians was saved. MRSA strains were kept at )80°C. Internal and external quality control using methicillin-susceptible bacteria (S. aureus ATCC 25923 = CIP 7625) and methicillinresistant bacteria (S. aureus CIP 6525 resistant to methicillin and susceptible to tobramycin and a MRSA resistant to tobramycin) was carried out by the laboratory on a regular basis. If a MRSA was identified in a nare or diagnostic specimen, the laboratory notified the vascular surgery unit immediately. Upon notification, pa-
Vol. 20, No. 6, 2006
tient carriers colonized and/or infected by MRSA benefited from septic, technical, and physical separation measures (for procedures and facilities); placement of a warning sticker at the entrance of the patientÕs room and on the care chart; and preventive measures for transfer within or outside the institution. These patients did not undergo treatment to eradicate carriage of MRSA such as mupirocin (BactrobanÒ 2% nasal ointment, Glaxo smith Kline, Marly-Le-Roi, France) or antiseptic cleansing. Study Data Microbiological, clinical, and epidemiological data were collected prospectively during the 4 months of study. The total number of screening specimens and of methicillin-susceptible S. aureus (MSSA) and MRSA strains isolated from specimens taken for screening and diagnostic purposes was determined. Epidemiological doublets, defined according to lÕObservatoire National de lÕEpide´miologie de la Re´sistance Bacte´rienne aux Antibiotiques (two strains isolated from the same patient during the study period are considered different on the basis of their antibiogram if there was a major discordance [resistant versus susceptible]), were excluded.5 The number of colonized and/or infected patients was determined. Colonization was defined as the presence of MRSA in the nares. Infection was defined as the presence of MRSA in a normally aseptic location in association with clinical or laboratory evidence of infection. A distinction was made between acquired and imported cases. The strain was considered imported if it was isolated in another department of the institution or in another institution or if it was present in the patient at the time of admission. The strain was considered acquired in the unit if it was isolated at least 48 hr after admission and if it was not present in the patient before admission. The following data were collected for all patients colonized and/or infected by MRSA: origin of the patient (i.e., from home, nonmedical facility, medical facility, or another care unit operated by the Tours Regional University Hospital center), hospitalization in the previous year, history of MRSA, antimicrobial therapy in the previous 3 months, skin lesions, and diabetes. Three French indicators recommended by the National Technical Committee on Nosocomial Infections (Ministry of Employment and Solidarity, Secretary of State for Health and Social Action, France) were calculated for strains isolated from specimens taken for diagnostic purposes. Those indicators are (1) percentage of methicillin
Nasal carriage of MRSA in vascular surgery 769
resistance in the S. aureus species, (2) incidence density of MRSA per 1,000 days of hospitalization, and (3) attack rate per 100 admissions per direct entry and entry via internal transfer within the same specialty group at the University Hospital Center. The statistical link between MRSA infection and MRSA colonization was analyzed by the chi-squared test. Analysis of Risk Factors A case-control study was used to analyze risk factors for MRSA carriage at admission to the vascular surgery unit. Cases were patient carriers of MRSA classified as imported (strains isolated at a later time in another unit and/or institution or isolated at admission to the unit). Cases were diagnosed during the study period (recruitment was therefore a confounding factor). Controls were patients without MRSA or MSSA, who were admitted to the unit during the study period. Each case was matched to four controls (type 1:4) on the basis of sex and age (±5 years). A total of eight risk factors defined based on the literature were studied. They were (1) origin of the patient from home or a nonmedical facility, (2) origin of the patient from another medical facility, (3) origin of the patient from another unit, (4) hospitalization in the previous year, (5) history of MRSA, (6) antimicrobial therapy within previous 3 months, (7) skin lesions, and (8) diabetes.6 Odds ratios (ORs) were estimated on an ad hoc basis and by 95% confidence intervals (95% CIs) using conditional logistic regression. Univariate analysis was done using SAS version 8.1 (SAS Institute, Cary, NC).
RESULTS MRSA Isolated from Nasal Screening Specimens During the 4-month study, a total of 468 swab samples were collected from 308 patients. Excluding epidemiological doublets, 13 nasal screening specimens were positive for MRSA; 4.2% of patients screened were MSRA carriers. In addition, 35 patients, i.e., 11.4% of patients screened, were identified as MSSA carriers. In the remaining 260 cases, specimens were negative for S. aureus (Fig. 1). The prevalence of methicillin resistance was 27.0% of S. aureus strains. The median age of patients colonized by MRSA was 73 years (range 43-89) and the sex ratio was 1.6 (eight men, five women). Three patients originated from their homes, seven from another medical facility, and five from another department
770 Morange-Saussier et al.
Annals of Vascular Surgery
35 ; 11.4% 100
13 ; 4.2%
80
9 293
60
(%)
40 20
260 ; 84.4%
4 2
0 Colonized MRSA
NEGATIVES
MRSA
MSSA
MRSA infections
Non colonized MRSA No MRSA infections
Fig. 1. Results of microbiological analysis of nasal specimens from the 308 patients screened.
Fig. 2. Infections due to MRSA in 13 patients with nasal colonization and 295 patients without nasal colonization.
of the University Hospital Center. Ten of the 13 MRSA carriers had been hospitalized in the previous 12 months. Only one of the 13 had a prior history of MRSA infection. Six of the 13 had received antimicrobial therapy in the previous 3 months. Nine of the 13 had skin lesions and five had diabetes. MRSA colonization was classified as imported in nine patients (69%) and acquired in four (31%) so that the rate of patient carriers screened at admission was 2.9% and the rate of MRSA acquisition in the unit was 1.3%. The median of MRSA acquisition in the nares was 7.5 days (range 4-11). Six patients underwent dirty surgery, five underwent clean surgery, and two had resolutive arterial ischemia without surgery. Four of the 13 carriers developed MRSA infection. Infection involved the stump area after lower limb amputation in two cases and femoropopliteal bypass in two.
incidence density was 3.9/1,000 days of hospitalization in the vascular surgery unit. The attack rate per 100 admissions was 1.9. The link between MRSA infection and colonization was highly significant (OR = 65, 95% CI 8.5-606, p < 0.001). The MRSA infection rate was 30.7% for patients with nasal colonization vs. 0.67% for patients without nasal colonization (Fig. 2).
MRSA Isolated from Specimens Taken for Diagnostic Purposes During the study period, 16 specimens taken for diagnostic purposes were positive for S. aureus. Six were methicillin-resistant, i.e., 37.5%. The other 10 specimens, i.e., 62.5%, were susceptible to methicillin. The sites of infection were surgical. Of the six patients infected by MRSA, four presented nasal colonization and two did not. Infection by MRSA was classified as imported in four patients (67%) and acquired in two patients (33%). Two of the four patients in whom MRSA infection was classified as imported had positive nasal carriage screening tests. The two patients who developed MRSA infection after 3 and 6 weeks of hospitalization presented nasal MRSA colonies at the time of admission. Five patients underwent dirty surgery and one underwent clean surgery. Over the 4-month study period, the percentage of MRSA in S. aureus strains was 43.7%. The
Case-Control Study to Evaluate Risk Factors for Carriage of MRSA at Admission to Vascular Surgery Unit A total of nine cases, i.e. patient carriers of a MRSA strain classified as imported, were included in the study. They were matched with 36 controls, i.e., patients without MRSA or MSSA. A significant correlation (p < 0.05) was observed between nasal carriage of MRSA and two variables, origin of patient from another department and origin of patient from another medical facility (Table I). A significant correlation was not observed between nasal carriage of MRSA and any of the other six variables studied, i.e., origin of patient from home or a nonmedical facility, hospitalization in the previous year, prior history of MRSA, antimicrobial therapy in the previous 3 months, skin lesions, and diabetes.
DISCUSSION Based on screening for 4 months, the prevalence of nasal carriage of MRSA in the vascular surgery unit was 4.2%. The MRSA acquisition rate in the vascular surgery unit was 1.3%, indicating that transmission by hand contact is low. MRSA is a public health problem due to spreading in hospital departments and increasing emergence of strains with reduced susceptibility to glycopeptides. Two types of measures are used to prevent spread of MRSA by cross-transmission,
Vol. 20, No. 6, 2006
Nasal carriage of MRSA in vascular surgery 771
Table I. Results of case-control study to evaluate risk factors for MRSA carriage at admission to the vascular surgery unit
Origin of patient Home Another department Another medical facility Hospitalization in previous year History of MRSA Antimicrobial therapy within previous 3 months Diabetes Skin lesions
OR
95% CI
p
1.00 15.59 20.24 2.00
1.14-212.94 2.20-186.23 0.13-30.41
0.039 0.008 0.618
4.00 4.74
0.25-63.95 0.87-25.85
0.327 0.072
1.00 4.41
0.24-4.11 0.48-40.36
1.000 0.189
early identification of carriers and isolation of carriers, which can include use of appropriate intraoperative antimicrobial prophylaxis. Nasal screening can detect MRSA carriers at admission who are not detected by specimens taken only for diagnostic purposes. S. aureus colonization of the nares is a risk factor for infection after cardiac7 and orthopedic8 surgery. MRSA cause 50% of SSIs involving S. aureus. In cardiac and orthopedic surgery, risk factors for nasal carriage of MRSA have been documented and presented at a French consensus conference on preoperative management of infectious risk on March 5, 2004. Guidelines include nasal screening for MRSA in patients admitted for cardiac or orthopedic surgery after mid- or long-term intensive care and in patients presenting chronic skin lesions. No French study has been conducted in vascular surgery units. Our study in the vascular surgery unit demonstrated that the prevalence of nasal carriage of MRSA is lower than in recent French studies. In 2004, Merrer et al.9 reported that the prevalence of nasal carriage of MRSA in a population of 179 patients benefiting from orthopedic surgery was 8.4%. In 2003, Lucet et al.6 reported a prevalence of 6.9% (range 3.7-20%) in multicenter studies on 2,347 patients hospitalized in intensive care units. Conversely, considerably lower prevalence of nasal carriage of MRSA has been reported, particularly in northern countries, e.g., 0.03% in a study involving 9,859 patients in the Netherlands,2 2% in the United States,10 and 2.6% in Turkey.11 Only four of the 13 patients colonized by MRSA had MRSA-positive diagnostic specimens. This suggests that screening identified nine MRSA cases that were not identified by specimens taken for
diagnostic purposes. The cost-effectiveness of nasal screening was high since we detected 60% of MRSA carriers who were unknown without routine nasal screening. These patients could be isolated to avoid spread of MRSA. This study also showed that nasal colonization by MRSA is a risk factor for MRSA infection since 30.7% of patients with MRSA colonization developed MRSA infection compared to only 0.67% of patients without MRSA colonization. These findings indicate that screening for nasal carriage of MRSA is useful in a vascular surgery unit. Should screening be performed routinely at admission or be targeted? Routine screening would have more impact than targeted screening, but it is work-intensive and costly. Targeted screening would have a strong impact on a few patients, but its overall impact would be limited. The risk group targeted for screening could be defined based on the two risk factors for MRSA carriage identified in our case-control study, i.e., origin of patient from another hospital department and origin of patient from another medical institution. These risk factors have already been reported in meta-analyses from the United States12 and Japan.13 The situation in vascular surgery appears to be similar to that in cardiac surgery, for which the risk factors for nasal carriage have been identified and documented.14 Like their counterparts undergoing cardiac surgery, patients scheduled for clean vascular surgery should be screened if they come from intensive care units, long- or mid-term care facilities, or departments with a high prevalence of MRSA or if they present chronic skin lesions. In our study, identification of MRSA colonization did not affect the surgical indication. Reports in the literature also appear to support the maintenance of surgical indications in patient carriers of MRSA. Two preventive strategies to reduce the risk of MRSA infection are currently being debated, i.e., decontamination using mupirocin and intraoperative antimicrobial prophylaxis in patient carriers of MRSA. Mupirocin was not used in any patient in our study. Decontamination of patient carriers using mupirocin (Bactroban 2% nasal ointment) is still controversial. Numerous studies have been carried out on eradicating S. aureus without demonstrating the efficacy of mupirocin on the incidence of SSI. In comparative studies published in 2002, Perl et al.15 and Kalmeijer et al.16 observed a decrease in SSI rate but the difference was not statistically significant. The reductions observed in three retrospective studies17-19 cannot be attributed to mupirocin alone due to the initial high SSI rate and
772 Morange-Saussier et al.
impact of appropriate control measures. Under these conditions, systematic use of mupirocin to prevent SSI in MRSA carriers cannot be recommended. French guidelines on the use of antimicrobial prophylaxis in surgery in 1999 recommend the use of vancomycin in patients with MRSA colonization.20 However, intraoperative antimicrobial prophylaxis has not been documented in terms of benefit for the patient (onset of SSI) and cost (development of vancomycin-resistant bacteria). Involvement of health-care personnel on a daily basis is critical to controlling the spread of AMB. To reduce transmission by hand contact, all medical staff must apply basic infection-control precautions and comply with separation measures for patients colonized or infected by AMB.
Annals of Vascular Surgery
8.
9.
10.
11.
12.
13.
REFERENCES 14. 1. Anonymous, Ministe`re de lÕemploi et de la solidarite´, Secre´tariat dÕEtat a` la Sante´ et a` lÕAction Sociale. Maıˆtrise de la diffusion des bacte´ries multire´sistantes aux antibiotiques. Paris: Ministe`re de lÕemploi et de la solidarite´. France, 1999. 2. Wertheim HFL, Vos MC, Boelens HAM, et al. Low prevalence of methicillin-resistant Staphylococcus aureus at hospital admission in the Netherlands: the value of search and destroy and restrictive antibiotic use. J Hosp Infect 2004; 56:321-325. 3. Brun Y, Bes M. Staphylococcus In: Freney, J, Renaud, F, Hansen, W, Bollet, C, Pre´cis de Bacte´riologie Clinique. Paris: ESKA, 2000, pp 783-830. 4. Communique´ 2004 du Comite´ de lÕAntibiogramme de la Socie´te´ Franc¸aise de Microbiologie, http://WWW.sfm.asso.fr. 5. Anonymous, ONERBA (Observatoire National de lÕEpide´miologie de la Re´sistance Bacte´rienne aux Antibiotiques). Rapport dÕactivite´ du conseil scientifique. Paris: ONERBA, 1998. 6. Lucet JC, Chevret S, Durand-Zaleski I, et al. Prevalence and risk factors for carriage of methicillin-resistant Staphylococcus aureus at admission to the intensive care unit. Arch Intern Med 2003;163:181-188. 7. Kluytmans JA, Mouton JW, Ijzerman EP, et al. Nasal carriage of Staphylococcus aureus as a major risk factor for wound
15.
16.
17.
18.
19.
20.
infections after cardiac surgery. J Infect Dis 1995;171:216219. Kalmeijer MD, van Nieuwland-Bollen PM, Bogaers-Hofman D, de Baere GA. Nasal carriage of Staphylococcus aureus is a major risk factor for surgical site infections in orthopaedic surgery. Infect Control Hosp Epidemiol 2000;21:319-323. Merrer J, Pisica-Donose G, Leneveu M, Pauthier F. Prevalence of methicillin-resistant Staphylococcus aureus nasal carriage among patient with femoral neck fractures: implication for antibiotic prophylaxis. Infect Control Hosp Epidemiol 2004;25:515-517. Leman R, Alvarado-Ramy F, Pocockb S, et al. Nasal carriage of methicillin-resistant Staphylococcus aureus in an American Indian population. Infect Control Hosp Epidemiol 2004;25:121-125. Cesur S, Cokca F. Nasal carriage of methicillin-resistant Staphylococcus aureus among hospital staff and outpatients. Infect Control Hosp Epidemiol 2004;25:169-171. Solgado CD, Farr BM, Calfee DP. Community-acquired methicillin-resistant Staphylococcus aureus: a meta-analysis of prevalence and risk factors. Clin Infect Dis 2003;36:131-139. Fukuda M, Tanaka H, Kajiwara Y, et al. High-risk populations for nasal carriage of methicillin-resistant Staphylococcus aureus. J Infect Chemother 2004;10:189-191. Confe´rence de consensus. Socie´te´ Franc¸aise d’Hygie`ne Hospitalie`re. Gestion pre´-ope´ratoire du risque infectieux. 5 mars 2004, Institut Pasteur Paris, http://WWW.sfhh.net. Perl TM, Cullen JJ, Wenzel RP, et al. Intranasal mupirocin to prevent postoperative Staphylococcus aureus infections. N Engl J Med 2002;346:1871-1877. Kalmeijer MD, Coertjens H, van Nieuwland-Bollen PM, et al. Surgical site infections in orthopedic surgery: the effect of mupirocin nasal ointment in a double-blind, randomized, placebo-controlled study. Clin Infect Dis 2002;35:353-358. Kluytmans JA, Mouton JW, VandenBergh MF, et al. Reduction of surgical-site infections in cardiothoracic surgery by elimination of nasal carriage of Staphylococcus aureus. Infect Control Hosp Epidemiol 1996;17:775-779. Cimochowski GE, Harostock MD, Brown R, et al. Intranasal mupirocin reduces sternal wound infection after open heart surgery in diabetics and nondiabetics. Ann Thorac Surg 2001;71:1572-1578. der Gernaat-van Sluis AJ, Hoogenboom-Verdegaal AM, Edixhoven PJ, Spies-van Rooijen NH. Prophylactic mupirocin could reduce orthopedic wound infections. 1,044 patients treated with mupirocin compared with 1,260 historical controls. Acta Orthop Scand 1998;69:412-414. Confe´rence de consensus de la Socie´te´ Franc¸aise dÕAnesthe´sie et de Re´animation. Antibioprophylaxie de lÕadulte en milieu chirurgical. Mars 1999, http://WWW.sfar.org/.
The purpose of this study was to determine the prevalence of nasal carriage of methicillinresistant Staphylococcus aureus (MRSA) and to define risk factors allowing identification of high-risk patients for MRSA nasal carriage at admission to the vascular surgery unit. From March 23, 2004 to July 13, 2004, screening for nasal carriage of MRSA was conducted at admission to the vascular surgery unit and 1 week thereafter. To analyze risk factors for MRSA nasal carriage at admission to the vascular surgery unit, a case-control study was carried out in patients presenting colonization at the time of admission. A total of 308 patients underwent nasal screening for MRSA. Thirteen were colonized with MRSA (nine at admission and four acquired), i.e., 4.2% of patients. Methicillin-susceptible Staphylococcus aureus (MSSA) was found in 11.4% of patients who underwent screening. Six patients with MRSA infection were identified during the study period. The two patients who acquired infection were colonized at the time of admission. Arrival from another health-care facility and from another department was a significant risk factor for carriage of MRSA. The prevalence of nasal carriage in vascular surgery was 4.2%. Nasal screening is highly cost-effective since 60% of MRSA carriers were undetected using diagnostic specimens alone. French recommendations issued for cardiac and orthopedic surgery by the consensus conference on preoperative management of infectious risk on March 5, 2004, should be extended to vascular surgery.
INTRODUCTION Antibiotic multiresistant bacteria (AMB) are a major cause of nosocomial infection (NI) in
1 Department of Bacteriology and Hospital Hygiene, Trousseau University Hospital, Tours, France. 2 INSERM CIC202, Franc¸ois Rabelais University and Regional University Hospital, Tours, France. 3 Vascular Surgery Department, Trousseau University Hospital, Tours, France. 4 EA 3854 ‘‘Bacteria and Materno-Fetal Risk,’’ Microbiology Department, University of Tours School of Medicine, Tours, France. Presented at the Twentieth Annual Meeting of the Socie´te´ de Chirurgie Vasculaire de Langue Franc¸aise, May 30-31, 2005, Lyon, France. Correspondence to: Virginie Morange-Saussier, MD, Bacteriology and Virology Department, Bretonneau Regional University Hospital Center, 37044, Tours, Cedex 1, France, E-mail: [email protected]
Ann Vasc Surg 2006; 20: 767–772 DOI: 10.1007/s10016-006-9088-x Ó Annals of Vascular Surgery Inc. Published online: November 4, 2006
France. The usual definition of NI is a hospitalacquired infection that was absent at the time of admission or developed within 48 hr after admission. Surgical site infections (SSIs) are considered nosocomial if they occur within 30 days after the intervention or, in cases involving implantation of a prosthesis or other device, within 1 year. Staphylococcus aureus is one of the two main bacterial species causing NI. Methicillin-resistant S. aureus (MRSA) is an AMB accounting for 5-10% of bacteria isolated during NI and 50% of S. aureus-related SSIs overall for all types of surgery.1 The multiresistance of these strains is defined as resistance to three antibiotics other than ß-lactamines, which limits the choice of antimicrobial therapy. Resistance of S. aureus to methicillin remains widespread in France and is involved in over 30% of all infections and 50% of NIs alone. High prevalence rates have also been documented in the United States (50%). A 767
768 Morange-Saussier et al.
previous study in 1990 showed that European countries could be divided along south and north lines with regard to the proportion of MRSA present in S. aureus strains, with Italy, France, and Spain showing around 30% and Denmark, Sweden, and the Netherlands showing around 2%.1 The lowest levels were observed in the Netherlands, with <1% of S. aureus found in clinical specimens being identified as MRSA.2 The 2001 national study of the prevalence of NI in France confirmed the high prevalence of AMB, i.e., 6.9% nationwide and 6.4% at the Tours University Hospital Center. MRSA accounted for 64% of S. aureus strains isolated from NI nationwide and 38% isolated at the Tours University Hospital. Whether symptomatic or not, infected or colonized carrier patients represent reservoirs for MRSA. They are a potential source of spreading and a major cause of epidemic MRSA outbreaks. Wound and urine infected with MRSA are important reservoirs. The nares, a preferential site of S. aureus colonization, are also an important reservoir. S. aureus colonizes the nares in 10-40% of individuals with no hospital contact. In the hospital setting and during epidemic outbreaks, up to 7% of the medical staff may be MRSA carriers.3 Transmission from carrier patients occurs primarily through the hands of health-care workers. However, transmission can also occur by contamination of the surface of inanimate objects such as stethoscopes and blood pressure cuffs. Screening and application of preventive measures for MRSA have been shown to be effective, especially in intensive care units. However, there are few data on other health-care settings, in particular for vascular surgery units.3 In this regard, it should be pointed out the risks associated with nasal carriage of MRSA have been confirmed for cardiac surgery, which presents conditions much like those observed in vascular surgery. A special feature of vascular surgery is to treat both septic and aseptic patients. The proximity between clean and contaminated patients raises a high risk of cross-transmission and therefore of catastrophic life-threatening prosthetic infection. By detecting MRSA carriers at admission, nasal screening may allow immediate implementation of control measures to prevent spreading. The purpose of this study was to establish the prevalence of MRSA carriage at admission to the vascular surgery unit and to evaluate the impact of eight risk factors for nasal carriage of MRSA in vascular surgery.
Annals of Vascular Surgery
MATERIALS AND METHODS Population This study included all patients hospitalized longer than 24 hr in a 19-bed vascular surgery unit over the 4-month period between March 23 and July 13, 2004. Microbiology Nurses took swab specimens from both of the nares of each patient at admission and 1 week thereafter. Patients who were readmitted were included. On the first day, nare specimens were seeded according to conventional culture technique on a selective Mueller-Hinton medium supplemented with tobramycin at 10 mg/L and on 5% nonselective blood agar gelose (bioMe´rieux, Marcy lÕEtoile, France). Gelose was then incubated for 24 hr at 37°C. On the second day, genus was diagnosed based on colony appearance, presence of a catalase, and characteristic microscopic morphology after gram staining (clustering of gram-positive cocci). S. aureus species identification was based on detection of bound coagulase by latex agglutination testing (PastorexÒ Staph-Plus; Bio-Rad, Ivry-sur-Seine, France). Speciation was confirmed biochemically using an automated Rapid ID32 STAPH system (bioMe´rieux). On the third day, susceptibility to methicillin was measured for each S. aureus strain (manual diffusion technique using a cefoxitin disk on gelose according to the recommendations of the Antibiogram Committee of the French Society of Microbiology in 2004).4 If necessary, methicillin resistance could be confirmed using either an oxacillin 5 mg disk incubated at 30°C for 24 hr, a rapid latex agglutination test (SlidexÒ MRSA detection, bioMe´rieux) to detect the PLP2A protein causing methicillin resistance, or molecular biology techniques to detect MecA coding for the PLP2A protein. MRSA from cultures ordered for diagnostic purposes by the clinicians was saved. MRSA strains were kept at )80°C. Internal and external quality control using methicillin-susceptible bacteria (S. aureus ATCC 25923 = CIP 7625) and methicillinresistant bacteria (S. aureus CIP 6525 resistant to methicillin and susceptible to tobramycin and a MRSA resistant to tobramycin) was carried out by the laboratory on a regular basis. If a MRSA was identified in a nare or diagnostic specimen, the laboratory notified the vascular surgery unit immediately. Upon notification, pa-
Vol. 20, No. 6, 2006
tient carriers colonized and/or infected by MRSA benefited from septic, technical, and physical separation measures (for procedures and facilities); placement of a warning sticker at the entrance of the patientÕs room and on the care chart; and preventive measures for transfer within or outside the institution. These patients did not undergo treatment to eradicate carriage of MRSA such as mupirocin (BactrobanÒ 2% nasal ointment, Glaxo smith Kline, Marly-Le-Roi, France) or antiseptic cleansing. Study Data Microbiological, clinical, and epidemiological data were collected prospectively during the 4 months of study. The total number of screening specimens and of methicillin-susceptible S. aureus (MSSA) and MRSA strains isolated from specimens taken for screening and diagnostic purposes was determined. Epidemiological doublets, defined according to lÕObservatoire National de lÕEpide´miologie de la Re´sistance Bacte´rienne aux Antibiotiques (two strains isolated from the same patient during the study period are considered different on the basis of their antibiogram if there was a major discordance [resistant versus susceptible]), were excluded.5 The number of colonized and/or infected patients was determined. Colonization was defined as the presence of MRSA in the nares. Infection was defined as the presence of MRSA in a normally aseptic location in association with clinical or laboratory evidence of infection. A distinction was made between acquired and imported cases. The strain was considered imported if it was isolated in another department of the institution or in another institution or if it was present in the patient at the time of admission. The strain was considered acquired in the unit if it was isolated at least 48 hr after admission and if it was not present in the patient before admission. The following data were collected for all patients colonized and/or infected by MRSA: origin of the patient (i.e., from home, nonmedical facility, medical facility, or another care unit operated by the Tours Regional University Hospital center), hospitalization in the previous year, history of MRSA, antimicrobial therapy in the previous 3 months, skin lesions, and diabetes. Three French indicators recommended by the National Technical Committee on Nosocomial Infections (Ministry of Employment and Solidarity, Secretary of State for Health and Social Action, France) were calculated for strains isolated from specimens taken for diagnostic purposes. Those indicators are (1) percentage of methicillin
Nasal carriage of MRSA in vascular surgery 769
resistance in the S. aureus species, (2) incidence density of MRSA per 1,000 days of hospitalization, and (3) attack rate per 100 admissions per direct entry and entry via internal transfer within the same specialty group at the University Hospital Center. The statistical link between MRSA infection and MRSA colonization was analyzed by the chi-squared test. Analysis of Risk Factors A case-control study was used to analyze risk factors for MRSA carriage at admission to the vascular surgery unit. Cases were patient carriers of MRSA classified as imported (strains isolated at a later time in another unit and/or institution or isolated at admission to the unit). Cases were diagnosed during the study period (recruitment was therefore a confounding factor). Controls were patients without MRSA or MSSA, who were admitted to the unit during the study period. Each case was matched to four controls (type 1:4) on the basis of sex and age (±5 years). A total of eight risk factors defined based on the literature were studied. They were (1) origin of the patient from home or a nonmedical facility, (2) origin of the patient from another medical facility, (3) origin of the patient from another unit, (4) hospitalization in the previous year, (5) history of MRSA, (6) antimicrobial therapy within previous 3 months, (7) skin lesions, and (8) diabetes.6 Odds ratios (ORs) were estimated on an ad hoc basis and by 95% confidence intervals (95% CIs) using conditional logistic regression. Univariate analysis was done using SAS version 8.1 (SAS Institute, Cary, NC).
RESULTS MRSA Isolated from Nasal Screening Specimens During the 4-month study, a total of 468 swab samples were collected from 308 patients. Excluding epidemiological doublets, 13 nasal screening specimens were positive for MRSA; 4.2% of patients screened were MSRA carriers. In addition, 35 patients, i.e., 11.4% of patients screened, were identified as MSSA carriers. In the remaining 260 cases, specimens were negative for S. aureus (Fig. 1). The prevalence of methicillin resistance was 27.0% of S. aureus strains. The median age of patients colonized by MRSA was 73 years (range 43-89) and the sex ratio was 1.6 (eight men, five women). Three patients originated from their homes, seven from another medical facility, and five from another department
770 Morange-Saussier et al.
Annals of Vascular Surgery
35 ; 11.4% 100
13 ; 4.2%
80
9 293
60
(%)
40 20
260 ; 84.4%
4 2
0 Colonized MRSA
NEGATIVES
MRSA
MSSA
MRSA infections
Non colonized MRSA No MRSA infections
Fig. 1. Results of microbiological analysis of nasal specimens from the 308 patients screened.
Fig. 2. Infections due to MRSA in 13 patients with nasal colonization and 295 patients without nasal colonization.
of the University Hospital Center. Ten of the 13 MRSA carriers had been hospitalized in the previous 12 months. Only one of the 13 had a prior history of MRSA infection. Six of the 13 had received antimicrobial therapy in the previous 3 months. Nine of the 13 had skin lesions and five had diabetes. MRSA colonization was classified as imported in nine patients (69%) and acquired in four (31%) so that the rate of patient carriers screened at admission was 2.9% and the rate of MRSA acquisition in the unit was 1.3%. The median of MRSA acquisition in the nares was 7.5 days (range 4-11). Six patients underwent dirty surgery, five underwent clean surgery, and two had resolutive arterial ischemia without surgery. Four of the 13 carriers developed MRSA infection. Infection involved the stump area after lower limb amputation in two cases and femoropopliteal bypass in two.
incidence density was 3.9/1,000 days of hospitalization in the vascular surgery unit. The attack rate per 100 admissions was 1.9. The link between MRSA infection and colonization was highly significant (OR = 65, 95% CI 8.5-606, p < 0.001). The MRSA infection rate was 30.7% for patients with nasal colonization vs. 0.67% for patients without nasal colonization (Fig. 2).
MRSA Isolated from Specimens Taken for Diagnostic Purposes During the study period, 16 specimens taken for diagnostic purposes were positive for S. aureus. Six were methicillin-resistant, i.e., 37.5%. The other 10 specimens, i.e., 62.5%, were susceptible to methicillin. The sites of infection were surgical. Of the six patients infected by MRSA, four presented nasal colonization and two did not. Infection by MRSA was classified as imported in four patients (67%) and acquired in two patients (33%). Two of the four patients in whom MRSA infection was classified as imported had positive nasal carriage screening tests. The two patients who developed MRSA infection after 3 and 6 weeks of hospitalization presented nasal MRSA colonies at the time of admission. Five patients underwent dirty surgery and one underwent clean surgery. Over the 4-month study period, the percentage of MRSA in S. aureus strains was 43.7%. The
Case-Control Study to Evaluate Risk Factors for Carriage of MRSA at Admission to Vascular Surgery Unit A total of nine cases, i.e. patient carriers of a MRSA strain classified as imported, were included in the study. They were matched with 36 controls, i.e., patients without MRSA or MSSA. A significant correlation (p < 0.05) was observed between nasal carriage of MRSA and two variables, origin of patient from another department and origin of patient from another medical facility (Table I). A significant correlation was not observed between nasal carriage of MRSA and any of the other six variables studied, i.e., origin of patient from home or a nonmedical facility, hospitalization in the previous year, prior history of MRSA, antimicrobial therapy in the previous 3 months, skin lesions, and diabetes.
DISCUSSION Based on screening for 4 months, the prevalence of nasal carriage of MRSA in the vascular surgery unit was 4.2%. The MRSA acquisition rate in the vascular surgery unit was 1.3%, indicating that transmission by hand contact is low. MRSA is a public health problem due to spreading in hospital departments and increasing emergence of strains with reduced susceptibility to glycopeptides. Two types of measures are used to prevent spread of MRSA by cross-transmission,
Vol. 20, No. 6, 2006
Nasal carriage of MRSA in vascular surgery 771
Table I. Results of case-control study to evaluate risk factors for MRSA carriage at admission to the vascular surgery unit
Origin of patient Home Another department Another medical facility Hospitalization in previous year History of MRSA Antimicrobial therapy within previous 3 months Diabetes Skin lesions
OR
95% CI
p
1.00 15.59 20.24 2.00
1.14-212.94 2.20-186.23 0.13-30.41
0.039 0.008 0.618
4.00 4.74
0.25-63.95 0.87-25.85
0.327 0.072
1.00 4.41
0.24-4.11 0.48-40.36
1.000 0.189
early identification of carriers and isolation of carriers, which can include use of appropriate intraoperative antimicrobial prophylaxis. Nasal screening can detect MRSA carriers at admission who are not detected by specimens taken only for diagnostic purposes. S. aureus colonization of the nares is a risk factor for infection after cardiac7 and orthopedic8 surgery. MRSA cause 50% of SSIs involving S. aureus. In cardiac and orthopedic surgery, risk factors for nasal carriage of MRSA have been documented and presented at a French consensus conference on preoperative management of infectious risk on March 5, 2004. Guidelines include nasal screening for MRSA in patients admitted for cardiac or orthopedic surgery after mid- or long-term intensive care and in patients presenting chronic skin lesions. No French study has been conducted in vascular surgery units. Our study in the vascular surgery unit demonstrated that the prevalence of nasal carriage of MRSA is lower than in recent French studies. In 2004, Merrer et al.9 reported that the prevalence of nasal carriage of MRSA in a population of 179 patients benefiting from orthopedic surgery was 8.4%. In 2003, Lucet et al.6 reported a prevalence of 6.9% (range 3.7-20%) in multicenter studies on 2,347 patients hospitalized in intensive care units. Conversely, considerably lower prevalence of nasal carriage of MRSA has been reported, particularly in northern countries, e.g., 0.03% in a study involving 9,859 patients in the Netherlands,2 2% in the United States,10 and 2.6% in Turkey.11 Only four of the 13 patients colonized by MRSA had MRSA-positive diagnostic specimens. This suggests that screening identified nine MRSA cases that were not identified by specimens taken for
diagnostic purposes. The cost-effectiveness of nasal screening was high since we detected 60% of MRSA carriers who were unknown without routine nasal screening. These patients could be isolated to avoid spread of MRSA. This study also showed that nasal colonization by MRSA is a risk factor for MRSA infection since 30.7% of patients with MRSA colonization developed MRSA infection compared to only 0.67% of patients without MRSA colonization. These findings indicate that screening for nasal carriage of MRSA is useful in a vascular surgery unit. Should screening be performed routinely at admission or be targeted? Routine screening would have more impact than targeted screening, but it is work-intensive and costly. Targeted screening would have a strong impact on a few patients, but its overall impact would be limited. The risk group targeted for screening could be defined based on the two risk factors for MRSA carriage identified in our case-control study, i.e., origin of patient from another hospital department and origin of patient from another medical institution. These risk factors have already been reported in meta-analyses from the United States12 and Japan.13 The situation in vascular surgery appears to be similar to that in cardiac surgery, for which the risk factors for nasal carriage have been identified and documented.14 Like their counterparts undergoing cardiac surgery, patients scheduled for clean vascular surgery should be screened if they come from intensive care units, long- or mid-term care facilities, or departments with a high prevalence of MRSA or if they present chronic skin lesions. In our study, identification of MRSA colonization did not affect the surgical indication. Reports in the literature also appear to support the maintenance of surgical indications in patient carriers of MRSA. Two preventive strategies to reduce the risk of MRSA infection are currently being debated, i.e., decontamination using mupirocin and intraoperative antimicrobial prophylaxis in patient carriers of MRSA. Mupirocin was not used in any patient in our study. Decontamination of patient carriers using mupirocin (Bactroban 2% nasal ointment) is still controversial. Numerous studies have been carried out on eradicating S. aureus without demonstrating the efficacy of mupirocin on the incidence of SSI. In comparative studies published in 2002, Perl et al.15 and Kalmeijer et al.16 observed a decrease in SSI rate but the difference was not statistically significant. The reductions observed in three retrospective studies17-19 cannot be attributed to mupirocin alone due to the initial high SSI rate and
772 Morange-Saussier et al.
impact of appropriate control measures. Under these conditions, systematic use of mupirocin to prevent SSI in MRSA carriers cannot be recommended. French guidelines on the use of antimicrobial prophylaxis in surgery in 1999 recommend the use of vancomycin in patients with MRSA colonization.20 However, intraoperative antimicrobial prophylaxis has not been documented in terms of benefit for the patient (onset of SSI) and cost (development of vancomycin-resistant bacteria). Involvement of health-care personnel on a daily basis is critical to controlling the spread of AMB. To reduce transmission by hand contact, all medical staff must apply basic infection-control precautions and comply with separation measures for patients colonized or infected by AMB.
Annals of Vascular Surgery
8.
9.
10.
11.
12.
13.
REFERENCES 14. 1. Anonymous, Ministe`re de lÕemploi et de la solidarite´, Secre´tariat dÕEtat a` la Sante´ et a` lÕAction Sociale. Maıˆtrise de la diffusion des bacte´ries multire´sistantes aux antibiotiques. Paris: Ministe`re de lÕemploi et de la solidarite´. France, 1999. 2. Wertheim HFL, Vos MC, Boelens HAM, et al. Low prevalence of methicillin-resistant Staphylococcus aureus at hospital admission in the Netherlands: the value of search and destroy and restrictive antibiotic use. J Hosp Infect 2004; 56:321-325. 3. Brun Y, Bes M. Staphylococcus In: Freney, J, Renaud, F, Hansen, W, Bollet, C, Pre´cis de Bacte´riologie Clinique. Paris: ESKA, 2000, pp 783-830. 4. Communique´ 2004 du Comite´ de lÕAntibiogramme de la Socie´te´ Franc¸aise de Microbiologie, http://WWW.sfm.asso.fr. 5. Anonymous, ONERBA (Observatoire National de lÕEpide´miologie de la Re´sistance Bacte´rienne aux Antibiotiques). Rapport dÕactivite´ du conseil scientifique. Paris: ONERBA, 1998. 6. Lucet JC, Chevret S, Durand-Zaleski I, et al. Prevalence and risk factors for carriage of methicillin-resistant Staphylococcus aureus at admission to the intensive care unit. Arch Intern Med 2003;163:181-188. 7. Kluytmans JA, Mouton JW, Ijzerman EP, et al. Nasal carriage of Staphylococcus aureus as a major risk factor for wound
15.
16.
17.
18.
19.
20.
infections after cardiac surgery. J Infect Dis 1995;171:216219. Kalmeijer MD, van Nieuwland-Bollen PM, Bogaers-Hofman D, de Baere GA. Nasal carriage of Staphylococcus aureus is a major risk factor for surgical site infections in orthopaedic surgery. Infect Control Hosp Epidemiol 2000;21:319-323. Merrer J, Pisica-Donose G, Leneveu M, Pauthier F. Prevalence of methicillin-resistant Staphylococcus aureus nasal carriage among patient with femoral neck fractures: implication for antibiotic prophylaxis. Infect Control Hosp Epidemiol 2004;25:515-517. Leman R, Alvarado-Ramy F, Pocockb S, et al. Nasal carriage of methicillin-resistant Staphylococcus aureus in an American Indian population. Infect Control Hosp Epidemiol 2004;25:121-125. Cesur S, Cokca F. Nasal carriage of methicillin-resistant Staphylococcus aureus among hospital staff and outpatients. Infect Control Hosp Epidemiol 2004;25:169-171. Solgado CD, Farr BM, Calfee DP. Community-acquired methicillin-resistant Staphylococcus aureus: a meta-analysis of prevalence and risk factors. Clin Infect Dis 2003;36:131-139. Fukuda M, Tanaka H, Kajiwara Y, et al. High-risk populations for nasal carriage of methicillin-resistant Staphylococcus aureus. J Infect Chemother 2004;10:189-191. Confe´rence de consensus. Socie´te´ Franc¸aise d’Hygie`ne Hospitalie`re. Gestion pre´-ope´ratoire du risque infectieux. 5 mars 2004, Institut Pasteur Paris, http://WWW.sfhh.net. Perl TM, Cullen JJ, Wenzel RP, et al. Intranasal mupirocin to prevent postoperative Staphylococcus aureus infections. N Engl J Med 2002;346:1871-1877. Kalmeijer MD, Coertjens H, van Nieuwland-Bollen PM, et al. Surgical site infections in orthopedic surgery: the effect of mupirocin nasal ointment in a double-blind, randomized, placebo-controlled study. Clin Infect Dis 2002;35:353-358. Kluytmans JA, Mouton JW, VandenBergh MF, et al. Reduction of surgical-site infections in cardiothoracic surgery by elimination of nasal carriage of Staphylococcus aureus. Infect Control Hosp Epidemiol 1996;17:775-779. Cimochowski GE, Harostock MD, Brown R, et al. Intranasal mupirocin reduces sternal wound infection after open heart surgery in diabetics and nondiabetics. Ann Thorac Surg 2001;71:1572-1578. der Gernaat-van Sluis AJ, Hoogenboom-Verdegaal AM, Edixhoven PJ, Spies-van Rooijen NH. Prophylactic mupirocin could reduce orthopedic wound infections. 1,044 patients treated with mupirocin compared with 1,260 historical controls. Acta Orthop Scand 1998;69:412-414. Confe´rence de consensus de la Socie´te´ Franc¸aise dÕAnesthe´sie et de Re´animation. Antibioprophylaxie de lÕadulte en milieu chirurgical. Mars 1999, http://WWW.sfar.org/.